From 300a7c11ce9614f53826870ad6ac415b837ba5e9 Mon Sep 17 00:00:00 2001
From: Hanhan Wang <hanchung@google.com>
Date: Tue, 12 Jan 2021 22:07:29 -0800
Subject: [PATCH] Upstream mhlo.dot_general lowering to Linalg to MHLO repo

PiperOrigin-RevId: 351514250
---
 .../mhlo/transforms/legalize_to_linalg.cc     | 86 ++++++++++++++++++-
 tests/hlo-legalize-to-linalg.mlir             | 21 +++++
 2 files changed, 106 insertions(+), 1 deletion(-)

diff --git a/lib/Dialect/mhlo/transforms/legalize_to_linalg.cc b/lib/Dialect/mhlo/transforms/legalize_to_linalg.cc
index f1cc090..9a31f8c 100644
--- a/lib/Dialect/mhlo/transforms/legalize_to_linalg.cc
+++ b/lib/Dialect/mhlo/transforms/legalize_to_linalg.cc
@@ -99,6 +99,14 @@ SmallVector<Value, 2> ExtractDynamicSizes(OpBuilder& b, Location loc,
   return dyn_sizes;
 }
 
+SmallVector<int64_t, 4> Extract1DVector(DenseIntElementsAttr elements) {
+  SmallVector<int64_t, 4> ret;
+  for (const APInt& element : elements) {
+    ret.push_back(element.getLimitedValue());
+  }
+  return ret;
+}
+
 template <typename OpTy, bool isLHLO = true>
 class PointwiseToLinalgConverter : public OpConversionPattern<OpTy> {
  public:
@@ -1131,6 +1139,81 @@ class DotOpOnTensorsConversion : public OpConversionPattern<mhlo::DotOp> {
   }
 };
 
+SmallVector<Value, 8> GetDotGeneralOpInitTensorDynSizes(
+    OpBuilder& b, Location loc, Value lhs, Value rhs, ShapedType result_type) {
+  SmallVector<Value, 8> dyn_shape;
+  if (result_type.isDynamicDim(0))
+    dyn_shape.push_back(b.create<DimOp>(loc, lhs, 0));
+  if (result_type.isDynamicDim(1))
+    dyn_shape.push_back(b.create<DimOp>(loc, lhs, 1));
+  if (result_type.isDynamicDim(2))
+    dyn_shape.push_back(b.create<DimOp>(loc, rhs, 2));
+  return dyn_shape;
+}
+
+class DotGeneralOpOnTensorsConversion
+    : public OpConversionPattern<mhlo::DotGeneralOp> {
+ public:
+  using OpConversionPattern<mhlo::DotGeneralOp>::OpConversionPattern;
+  LogicalResult matchAndRewrite(
+      mhlo::DotGeneralOp op, ArrayRef<Value> args,
+      ConversionPatternRewriter& rewriter) const final {
+    if (!VerifyHloOpBufferOrTensorSemantics</*isLHLO=*/false>(op)) {
+      return failure();
+    }
+    mhlo::DotDimensionNumbers dim_numbers = op.dot_dimension_numbers();
+    auto lhs_bathcing_dims =
+        Extract1DVector(dim_numbers.lhs_batching_dimensions());
+    auto rhs_bathcing_dims =
+        Extract1DVector(dim_numbers.rhs_batching_dimensions());
+    auto lhs_contracting_dims =
+        Extract1DVector(dim_numbers.lhs_contracting_dimensions());
+    auto rhs_contracting_dims =
+        Extract1DVector(dim_numbers.rhs_contracting_dimensions());
+    if (lhs_bathcing_dims.size() != 1 || lhs_bathcing_dims[0] != 0) {
+      return rewriter.notifyMatchFailure(
+          op, "expected lhs batching dimensions exactly {0}");
+    }
+    if (rhs_bathcing_dims.size() != 1 || rhs_bathcing_dims[0] != 0) {
+      return rewriter.notifyMatchFailure(
+          op, "expected rhs batching dimensions exactly {0}");
+    }
+    if (lhs_contracting_dims.size() != 1 || lhs_contracting_dims[0] != 2) {
+      return rewriter.notifyMatchFailure(
+          op, "expected lhs contracting dimensions exactly {2}");
+    }
+    if (rhs_contracting_dims.size() != 1 || rhs_contracting_dims[0] != 1) {
+      return rewriter.notifyMatchFailure(
+          op, "expected rhs contracting dimensions exactly {1}");
+    }
+    Location loc = op.getLoc();
+    mhlo::DotGeneralOp::Adaptor adaptor(args);
+    Type result_type = op.getResult().getType();
+    auto shaped_type = result_type.cast<ShapedType>();
+    SmallVector<Value, 8> dyn_shape = GetDotGeneralOpInitTensorDynSizes(
+        rewriter, loc, adaptor.lhs(), adaptor.rhs(), shaped_type);
+    auto zero_attr = rewriter.getZeroAttr(shaped_type.getElementType());
+    Value zero = rewriter.create<ConstantOp>(loc, zero_attr);
+    auto init_tensor = rewriter.create<DynamicTensorFromElementsOp>(
+        loc, result_type, dyn_shape);
+    {
+      OpBuilder::InsertionGuard guard(rewriter);
+      SmallVector<Type, 4> arg_types(shaped_type.getRank(),
+                                     rewriter.getIndexType());
+      Region& region = init_tensor.body();
+      Block* block = rewriter.createBlock(&region, region.begin(), arg_types);
+      rewriter.setInsertionPointToEnd(block);
+      rewriter.create<YieldOp>(loc, zero);
+    }
+    auto linalg_op = rewriter.create<linalg::BatchMatmulOp>(
+        loc, /*resultTensorTypes=*/TypeRange{result_type},
+        /*inputs=*/ValueRange{adaptor.lhs(), adaptor.rhs()},
+        /*outputBuffers=*/ValueRange{init_tensor});
+    rewriter.replaceOp(op, linalg_op.getResults());
+    return success();
+  }
+};
+
 void populateLHLOToLinalgConversionPattern(MLIRContext* context,
                                            OwningRewritePatternList* patterns) {
   // clang-format off
@@ -1299,7 +1382,8 @@ void populateHLOToLinalgConversionPattern(MLIRContext* context,
                ReshapeOpConverter<mhlo::ReshapeOp, false>,
                ReverseConverter<mhlo::ReverseOp, false>,
                TransposeConverter<mhlo::TransposeOp, false>,
-               DotOpOnTensorsConversion>(context);
+               DotOpOnTensorsConversion, DotGeneralOpOnTensorsConversion>(
+          context);
 }
 
 std::unique_ptr<OperationPass<FuncOp>> createLegalizeHloToLinalgPass() {
diff --git a/tests/hlo-legalize-to-linalg.mlir b/tests/hlo-legalize-to-linalg.mlir
index b5d762e..0058307 100644
--- a/tests/hlo-legalize-to-linalg.mlir
+++ b/tests/hlo-legalize-to-linalg.mlir
@@ -871,3 +871,24 @@ func @dot_dot(%arg0: tensor<?xf32>,
 // CHECK: linalg.dot
 // CHECK-SAME: ins(%[[ARG0]], %[[ARG1]] : tensor<?xf32>, tensor<?xf32>)
 // CHECK-SAME: outs(%[[INIT]] : tensor<f32>)
+
+// -----
+
+func @dot_general(%arg0: tensor<?x?x3xf32>,
+                  %arg1: tensor<?x3x?xf32>) -> tensor<?x?x?xf32> {
+  %0 ="mhlo.dot_general"(%arg0, %arg1) {
+      dot_dimension_numbers = {
+          lhs_batching_dimensions = dense<0> : tensor<1xi64>,
+          lhs_contracting_dimensions = dense<2> : tensor<1xi64>,
+          rhs_batching_dimensions = dense<0> : tensor<1xi64>,
+          rhs_contracting_dimensions = dense<1> : tensor<1xi64>
+      },
+      precision_config = ["DEFAULT", "DEFAULT"]
+  } : (tensor<?x?x3xf32>, tensor<?x3x?xf32>) -> tensor<?x?x?xf32>
+  return %0 : tensor<?x?x?xf32>
+}
+// CHECK: func @dot_general(%[[ARG0:.*]]: tensor<?x?x3xf32>, %[[ARG1:.*]]: tensor<?x3x?xf32>)
+// CHECK: %[[INIT:.*]] = dynamic_tensor_from_elements
+// CHECK: linalg.batch_matmul
+// CHECK-SAME: ins(%[[ARG0]], %[[ARG1]] : tensor<?x?x3xf32>, tensor<?x3x?xf32>)
+// CHECK-SAME: outs(%[[INIT]] : tensor<?x?x?xf32>)